CN108181808A - Inclined methods of self-tuning of the form Non-Model Controller based on systematic error of MISO - Google Patents

Inclined methods of self-tuning of the form Non-Model Controller based on systematic error of MISO Download PDF

Info

Publication number
CN108181808A
CN108181808A CN201711323412.9A CN201711323412A CN108181808A CN 108181808 A CN108181808 A CN 108181808A CN 201711323412 A CN201711323412 A CN 201711323412A CN 108181808 A CN108181808 A CN 108181808A
Authority
CN
China
Prior art keywords
miso
systematic error
form non
moment
model
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201711323412.9A
Other languages
Chinese (zh)
Other versions
CN108181808B (en
Inventor
卢建刚
李雪园
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN201711323412.9A priority Critical patent/CN108181808B/en
Publication of CN108181808A publication Critical patent/CN108181808A/en
Application granted granted Critical
Publication of CN108181808B publication Critical patent/CN108181808B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention discloses a kind of inclined methods of self-tuning of the form Non-Model Controller based on systematic error of MISO, by the use of systematic error collection as the input of BP neural network, BP neural network carries out forward calculation and passes through output layer and export penalty factor, the inclined form Non-Model Controllers of the MISO such as step factor treat setting parameter, the control input vector for controlled device is calculated using the control algolithm of the inclined form Non-Model Controllers of MISO, target is minimised as with the value of system error function, using gradient descent method, and it combines control input and is directed to each gradient information collection for treating setting parameter respectively, carry out systematic error backpropagation calculating, the hidden layer weight coefficient of online real-time update BP neural network, output layer weight coefficient, realize parameter self-tuning of the controller based on systematic error.Inclined methods of self-tuning of the form Non-Model Controller based on systematic error of MISO proposed by the present invention can effectively overcome the on-line tuning problem of controller parameter, have good control effect to MISO system.

Description

Inclined methods of self-tuning of the form Non-Model Controller based on systematic error of MISO
Technical field
The invention belongs to automation control areas, and system is based on more particularly, to a kind of inclined form Non-Model Controllers of MISO The methods of self-tuning of error.
Background technology
The control problem of MISO (Multiple Input and Single Output, multiple input single output) system, one One of significant challenge that control field is faced has been automated since straight.
The existing implementation method of MISO controllers includes the inclined form Non-Model Controllers of MISO.The inclined form model-frees of MISO Controller is a kind of novel data drive control method, does not depend on any mathematical model information of controlled device, only relies upon The inputoutput data that MISO controlled devices measure in real time into line control unit analysis and design, and realize it is concise, calculate it is negative Small and strong robustness is carried on a shoulder pole, unknown nonlinear time-varying MISO system can also be controlled well, before there is good application Scape.The theoretical foundation of the inclined form Non-Model Controllers of MISO is collaborateed by Hou Zhong lifes with Jin Shangtai at it《Model-free adaption Control-theory and application》It is proposed in (Science Press, 2013, page 106), control algolithm is as follows：
Wherein, control input vectors of the u (k) for the k moment, u (k)=[u1(k),…,um(k)]T, m inputs a in order to control Number, Δ u (k)=u (k)-u (k-1)；E (k) is the systematic error at k moment；MISO system puppet piecemeal gradient for the k moment The row matrix of estimated value,For row matrixI-th piece of row matrix (i=1 ..., L),For row matrix 2 norms；λ is penalty factor, ρ1,…,ρLFor step factor, L input linear length constants in order to control.
However, the inclined form Non-Model Controllers of MISO need to rely on Heuristics before actually coming into operation punishment is previously set Factor lambda and step factor ρ1,…,ρLEtc. parameters numerical value, penalty factor λ and step-length are also not yet realized during actually come into operation Factor ρ1,…,ρLEtc. parameters online self-tuning.Parameter effectively adjusts the shortage of means, not only makes the inclined form model-free controls of MISO The use debugging process of device processed is time-consuming and laborious, and can also seriously affect the control effect of the inclined form Non-Model Controllers of MISO sometimes Fruit constrains the popularization and application of the inclined form Non-Model Controllers of MISO.That is：The inclined form Non-Model Controllers of MISO are in reality Border also needs to solve the problems, such as online self-tuning parameter during coming into operation.
For this purpose, in order to break the bottleneck for restricting the inclined form Non-Model Controllers of MISO and promoting and applying, the present invention proposes one Kind inclined methods of self-tuning of the form Non-Model Controller based on systematic error of MISO.
Invention content
In order to solve the problems, such as background technology, the object of the present invention is to provide a kind of inclined forms of MISO without mould Methods of self-tuning of the type controller based on systematic error.
For this purpose, the above-mentioned purpose of the present invention is achieved through the following technical solutions, include the following steps：
Step (1)：For with MISO (Multiple of the m input (m is the integer more than or equal to 2) with 1 output Input and Single Output, multiple input single output) system, it is controlled using the inclined form Non-Model Controllers of MISO； Control input linear the length constant L, L for determining the inclined form Non-Model Controllers of the MISO are the integer more than 1；It is described The inclined form Non-Model Controller parameters of MISO include penalty factor λ and step factor ρ1,…,ρL；Determine the inclined form model-frees of MISO Controller treats setting parameter, and the inclined form Non-Model Controllers of MISO treat setting parameter, is the inclined form model-frees of the MISO Controller parameter it is part or all of, include penalty factor λ and step factor ρ1,…,ρLOne of arbitrary or any number of combination； Determine the input layer number, node in hidden layer, output layer number of nodes of BP neural network, the output layer number of nodes is no less than The inclined form Non-Model Controllers of MISO treat setting parameter number；Initialize the BP neural network hidden layer weight coefficient, Output layer weight coefficient；
Step (2)：The k moment will be denoted as current time；
Step (3)：Actual value exported based on system output desired value and system, function is calculated using systematic error To the systematic error at k moment, it is denoted as e (k)；By the systematic error and its group of functions, system output desired value, system output One of the arbitrary or any number of combination of actual value, is put into set { systematic error collection }；
Step (4)：The set { systematic error collection } that step (3) is obtained is described as the input of BP neural network BP neural network carries out forward calculation, and result of calculation is exported by the output layer of the BP neural network, and it is inclined to obtain the MISO Form Non-Model Controller treats the value of setting parameter；
Step (5)：The inclined forms of the MISO that the systematic error e (k) that is obtained based on step (3), step (4) are obtained Non-Model Controller treats the value of setting parameter, and using the control algolithm of the inclined form Non-Model Controllers of MISO, MISO is calculated Inclined form Non-Model Controller is directed to control input vector u (k)=[u of the controlled device at the k moment1(k),…,um(k)]T；
Step (6)：J-th of control input u in the control input vector u (k) obtained for step (5)j(k) (1≤j≤m) calculates j-th of the control input uj(k) it is treated respectively for each inclined form Non-Model Controllers of MISO Gradient information of the setting parameter at the k moment, specific formula for calculation are as follows：
When the inclined form Non-Model Controllers of the MISO when in setting parameter include penalty factor λ when, it is described j-th control Input uj(k) it is for gradient informations of the penalty factor λ at the k moment：
When the inclined form Non-Model Controllers of the MISO are treated to include step factor ρ in setting parameter1When, j-th of control System input uj(k) for the step factor ρ1It is in the gradient information at k moment：
When the inclined form Non-Model Controllers of the MISO are treated to include step factor ρ in setting parameteriAnd during 2≤i≤L, institute State jth control input uj(k) for the step factor ρiIt is in the gradient information at k moment：
Wherein, Δ uj(k)=uj(k)-uj(k-1),The row of MISO system puppet piecemeal gradient estimated value for the k moment Matrix,For row matrixI-th piece of row matrix (i=1 ..., L),For row matrixJ-th of gradient Component estimated value,For row matrix2 norms；
The set of the above-mentioned whole gradient information is denoted as { gradient information j }, is put into set { gradient information collection }；
Other m-1 control input in the control input vector u (k) obtained for step (5), repeats this Step, until the set { gradient information collection } includes the set of all { { gradient information 1 } ..., { gradient information m } }, then It enters step (7)；
Step (7)：Target is minimised as with the value of system error function, using gradient descent method, is obtained with reference to step (6) The set { gradient information collection }, carry out systematic error backpropagation calculating, update BP neural network hidden layer weight coefficient, Output layer weight coefficient, as later moment in time BP neural network carry out forward calculation when hidden layer weight coefficient, output layer weight coefficient；
Step (8)：After the control input vector u (k) acts on controlled device, controlled device is obtained in later moment in time System exports actual value, back to step (2), repeats step (2) to step (8).
While using above-mentioned technical proposal, the present invention can also be used or be combined using technology further below Scheme：
It is defeated comprising system output desired value and system that the systematic error in the step (3) calculates argument of function Go out actual value.
The systematic error in the step (3) calculates function and uses e (k)=y*(k)-y (k), wherein y*(k) for k when The system output desired value of setting is carved, y (k) is that the system that k instance samples obtain exports actual value；Or using e (k)=y*(k + 1)-y (k), wherein y*(k+1) desired value is exported for the system at k+1 moment, y (k) is that the system output that k instance samples obtain is real Actual value.
The systematic error and its group of functions in the step (3), the systematic error e (k) comprising the k moment, the k moment and The accumulation of the systematic error at all moment before isSingle order backward difference e (k)-e (k- of moment systematic error e (k) 1), the height of second order backward difference e (k) -2e (k-1)+e (k-2) of k moment systematic error e (k), k moment systematic error e (k) One of the arbitrary or any number of combination of rank backward difference.
The independent variable of the system error function in the step (7) includes systematic error, system output desired value, is One of the arbitrary or any number of combination of system output actual value.
The system error function in the step (7) isWherein, e (k) is missed for system Difference, Δ uj(k)=uj(k)-uj(k-1), bjTo be greater than or equal to 0 constant, 1≤j≤m.
Inclined methods of self-tuning of the form Non-Model Controller based on systematic error of MISO provided by the invention, Neng Goushi Existing good control effect, and effectively overcome penalty factor λ and step factor ρ1,…,ρLNeed the time-consuming and laborious difficulty adjusted Topic.
Description of the drawings
Fig. 1 is the principle of the present invention block diagram；
Fig. 2 is the BP neural network structure diagram that the present invention uses；
Fig. 3 is the single output MISO system of two inputs in penalty factor λ and step factor ρ1,ρ2,ρ3Control during Self-tuning System simultaneously Design sketch processed；
Fig. 4 is the single output MISO system of two inputs in penalty factor λ and step factor ρ1,ρ2,ρ3Control during Self-tuning System simultaneously System input figure；
Fig. 5 is the single output MISO system of two inputs in penalty factor λ and step factor ρ1,ρ2,ρ3Punishing during Self-tuning System simultaneously Penalty factor λ change curves；
Fig. 6 is the single output MISO system of two inputs in penalty factor λ and step factor ρ1,ρ2,ρ3Step during Self-tuning System simultaneously Long factor ρ1,ρ2,ρ3Change curve；
Fig. 7 is fixed and step factor ρ for the single output MISO system of two inputs in penalty factor λ1,ρ2,ρ3Control during Self-tuning System Design sketch processed；
Fig. 8 is fixed and step factor ρ for the single output MISO system of two inputs in penalty factor λ1,ρ2,ρ3Control during Self-tuning System System input figure；
Fig. 9 is fixed and step factor ρ for the single output MISO system of two inputs in penalty factor λ1,ρ2,ρ3Step during Self-tuning System Long factor ρ1,ρ2,ρ3Change curve.
Specific embodiment
The present invention is further described in the following with reference to the drawings and specific embodiments.
Fig. 1 gives the principle of the present invention block diagram.For with m input (m is the integer more than or equal to 2) and 1 The MISO system of output is controlled using the inclined form Non-Model Controllers of MISO；Determine the inclined form Non-Model Controllers of MISO Control input linear length constant L, L is the integer more than 1；The inclined form Non-Model Controller parameters of MISO include punishment because Sub- λ and step factor ρ1,…,ρL；It determines that the inclined form Non-Model Controllers of MISO treat setting parameter, is the inclined forms of the MISO Non-Model Controller parameter it is part or all of, include penalty factor λ and step factor ρ1,…,ρLIt is one of arbitrary or any number of Combination；In Fig. 1, the inclined form Non-Model Controllers of MISO treat that setting parameter is penalty factor λ and step factor ρ1,…,ρL；Really Determine the input layer number, node in hidden layer, output layer number of nodes of BP neural network, wherein output layer number of nodes is no less than The inclined form Non-Model Controllers of MISO treat setting parameter number；Initialize hidden layer weight coefficient, the output of the BP neural network Layer weight coefficient.
The k moment will be denoted as current time；By system output desired value y*(k) and system output actual value y (k) difference conduct The systematic error e (k) at k moment, then by the systematic error e (k) at k moment, k moment and the systematic error at all moment before Accumulation isThe combination of single order backward difference e (k)-e (k-1) of moment systematic error e (k), is put into set { system Error collection }；The input of { systematic error collection } as BP neural network will be gathered, BP neural network carries out forward calculation, calculates knot Fruit is exported by the output layer of BP neural network, obtains the value that the inclined form Non-Model Controllers of MISO treat setting parameter；Based on institute The value that the inclined form Non-Model Controller of systematic error e (k), the MISO treats setting parameter is stated, using the inclined form model-frees of MISO The control algolithm of controller is calculated the inclined form Non-Model Controllers of MISO and is inputted for control of the controlled device at the k moment Vectorial u (k)=[u1(k),…,um(k)]T；For j-th of control input u in control input vector u (k)j(k) (1≤j≤ M), j-th of the control input u is calculatedj(k) setting parameter is treated for each inclined form Non-Model Controllers of MISO respectively { gradient information j } is denoted as in the gradient information at k moment, and by the set of all gradient informations, is put into set { gradient information Collection }；For other m-1 control input in control input vector u (k), repeat until set { gradient information collection } packet Set containing all { { gradient information 1 } ..., { gradient information m } }；Then, with reference to the set { gradient information collection }, with system The value of error function is minimised as target, with e in Fig. 12(k) target is minimised as, using gradient descent method, carries out systematic error Backpropagation calculates, and hidden layer weight coefficient, the output layer weight coefficient of BP neural network is updated, as later moment in time BP nerve nets Hidden layer weight coefficient, output layer weight coefficient during network progress forward calculation；After control input vector u (k) acts on controlled device, The system that controlled device is obtained in later moment in time exports actual value, then repeats the work described in this paragraph, carries out latter The parameter self-tuning process of the inclined form Non-Model Controllers of MISO based on systematic error at moment.
Fig. 2 gives the BP neural network structure diagram that the present invention uses.Hidden layer may be used in BP neural network The structure of individual layer can also use structure of the hidden layer for multilayer.In the schematic diagram of Fig. 2, for simplicity, BP neural network The structure that hidden layer is individual layer is employed, that is, uses the Three Tiered Network Architecture being made of input layer, individual layer hidden layer, output layer, Input layer number is set as 3, and node in hidden layer is set as 6, and output layer number of nodes is set as treating setting parameter number (in Fig. 2 Treat setting parameter number for L+1).3 nodes of input layer, with systematic error e (k), the accumulation of systematic errorSystem Single order backward difference e (k)-e (k-1) of system error e (k) is corresponded to respectively.The node of output layer, with penalty factor λ and step-length because Sub- ρ1,…,ρLIt corresponds to respectively.The hidden layer weight coefficient of BP neural network, the renewal process of output layer weight coefficient are specially：To be The value of system error function is minimised as target, with e in Fig. 22(k) target is minimised as, using gradient descent method, with reference to the collection It closes { gradient information collection }, carries out systematic error backpropagation calculating, so as to update the hidden layer weight coefficient of BP neural network, output Layer weight coefficient.
It is the specific embodiment of the present invention below.
Two inputs single output MISO system of the controlled device for typical non linear：
System output desired value y*(k) it is as follows：
y*(k)=(- 1)round((k-1)/100)
In this embodiment, m=2.
The numerical value of the control input linear length constant L of the inclined form Non-Model Controllers of MISO is generally according to controlled device Complexity and practical control effect set, it is too small to influence control effect generally between 1 to 10, cross conference Cause computationally intensive, so generally often taking 3 or 5, L is taken as 3 in this embodiment.
BP neural network uses the Three Tiered Network Architecture being made of input layer, individual layer hidden layer, output layer, input layer Number is set as 3, and node in hidden layer is set as 6, and output layer number of nodes is set as treating setting parameter number.
For above-mentioned specific embodiment, two groups of verification experimental verifications have been carried out altogether.
During first group of verification experimental verification, the output layer number of nodes of BP neural network is preset as 4 in Fig. 2, to penalty factor λ and Step factor ρ1,ρ2,ρ3Self-tuning System simultaneously is carried out, design sketch, Fig. 4 input figure to Fig. 3 in order to control in order to control, and Fig. 5 is penalty factor λ Change curve, Fig. 6 are step factor ρ1,ρ2,ρ3Change curve.The result shows that method of the invention by penalty factor λ and Step factor ρ1,ρ2,ρ3Self-tuning System simultaneously is carried out, can realize good control effect, and can effectively overcome penalty factor λ and step factor ρ1,ρ2,ρ3Need the time-consuming and laborious problem adjusted.
During second group of verification experimental verification, the output layer number of nodes of BP neural network is 3 in Fig. 2, first consolidates penalty factor λ The average value of penalty factor λ when determining value for first group of verification experimental verification, then to step factor ρ1,ρ2,ρ3Carry out Self-tuning System, figure 7 design sketch in order to control, Fig. 8 input figure in order to control, and Fig. 9 is step factor ρ1,ρ2,ρ3Change curve.As a result it again shows that, this hair Bright method is when penalty factor λ is fixed by step factor ρ1,ρ2,ρ3Self-tuning System is carried out, can realize good control effect Fruit, and can effectively overcome step factor ρ1,ρ2,ρ3Need the time-consuming and laborious problem adjusted.
System should be exported into desired value y it is emphasized that in above-mentioned specific embodiment*(k) it is real with system output The difference of actual value y (k) is as systematic error e (k), that is, e (k)=y*(k)-y (k), only described systematic error calculate function In a kind of method；The system at k+1 moment can also be exported desired value y*(k+1) actual value y is exported with the system at k moment (k) difference is as systematic error e (k), that is, e (k)=y*(k+1)-y(k)；The systematic error calculates function and can also adopt Other computational methods of system output desired value and system output actual value are included with independent variable, for example, For the controlled device of above-mentioned specific embodiment, missed using above-mentioned different system Difference calculates function, can realize good control effect.
Should also be it is emphasized that in above-mentioned specific embodiment, the set inputted as BP neural network { miss by system Difference set } select the accumulation of systematic error e (k), systematic errorThe single order backward difference e (k) of systematic error e (k)- The combination of e (k-1), only one of which are combined；The set { systematic error collection } can also use other combinations, and citing comes It says, the accumulation for systematic error e (k), systematic error isSingle order backward difference e (k)-e (k- of systematic error e (k) 1), second order backward difference e (k) -2e (k-1)+e (k-2) of systematic error e (k), the third-order and fourthorder of systematic error e (k) or more One of the arbitrary or any number of combination of the functions such as the backward difference of high-order.For the controlled device of above-mentioned specific embodiment, adopt With above-mentioned different set { systematic error collection }, good control effect can be realized.
More mesh should be minimised as with the value of system error function it is emphasized that in above-mentioned specific embodiment When marking hidden layer weight coefficient, the output layer weight coefficient to update BP neural network, the system error function uses e2(k), only For a kind of function in the system error function；The system error function can also use independent variable include systematic error, System output desired value, system output actual value one of arbitrary or any number of combination other functions, for example, system is missed Difference function uses (y*(k)-y(k))2Or (y*(k+1)-y(k))2, that is, using e2(k) another functional form；It illustrates again For, system error function usesWherein, Δ uj(k)=uj(k)-uj(k-1), bjTo be more than or Constant equal to 0,1≤j≤m；Obviously, work as bjWhen being equal to 0, system error function only accounts for e2(k) contribution, shows most The target of smallization is systematic error minimum, that is, pursues precision height；And work as bjDuring more than 0, system error function considers simultaneously e2(k) contribution andContribution, show minimize target pursue systematic error it is small while, also pursue control it is defeated Enter to change small, that is, not only pursued high pursue again of precision and manipulated surely.For the controlled device of above-mentioned specific embodiment, in use Different system error functions is stated, can realize good control effect；Only consider e with system error function2(k) when contributing Control effect compare, consider e simultaneously in system error function2(k) contribution andContribution when its control accuracy omit Having reduces and it manipulates stationarity and is then improved.
Finally should it is emphasized that the inclined form Non-Model Controllers of the MISO treat setting parameter, comprising punishment because Sub- λ and step factor ρ1,…,ρLOne of arbitrary or any number of combination；In above-mentioned specific embodiment, first group of verification experimental verification When penalty factor λ and step factor ρ1,ρ2,ρ3Realize while Self-tuning System, during second group of verification experimental verification penalty factor λ fix and Step factor ρ1,ρ2,ρ3Realize Self-tuning System；In practical application, it can also select to treat setting parameter as the case may be Any number of combination, for example, step factor ρ1,ρ2It fixes and penalty factor λ, step factor ρ3Realize Self-tuning System；In addition, The inclined form Non-Model Controllers of MISO treat setting parameter, including but not limited to penalty factor λ and step factor ρ1,…,ρL, citing For, as the case may be, the row matrix of MISO system puppet piecemeal gradient estimated value can also be includedEtc. parameters.
Above-mentioned specific embodiment is used for illustrating the present invention, is merely a preferred embodiment of the present invention rather than to this Invention is limited, and in the protection domain of spirit and claims of the present invention, to any modification of the invention made, is equal Replace, improve etc., both fall within protection scope of the present invention.

Claims (6)

  1. Inclined methods of self-tuning of the form Non-Model Controller based on systematic error of 1.MISO, it is characterised in that including following step Suddenly：
    Step (1)：For with MISO (Multiple of the m input (m is the integer more than or equal to 2) with 1 output Input and Single Output, multiple input single output) system, it is controlled using the inclined form Non-Model Controllers of MISO； Control input linear the length constant L, L for determining the inclined form Non-Model Controllers of the MISO are the integer more than 1；It is described The inclined form Non-Model Controller parameters of MISO include penalty factor λ and step factor ρ1,…,ρL；Determine the inclined form model-frees of MISO Controller treats setting parameter, and the inclined form Non-Model Controllers of MISO treat setting parameter, is the inclined form model-frees of the MISO Controller parameter it is part or all of, include penalty factor λ and step factor ρ1,…,ρLOne of arbitrary or any number of combination； Determine the input layer number, node in hidden layer, output layer number of nodes of BP neural network, the output layer number of nodes is no less than The inclined form Non-Model Controllers of MISO treat setting parameter number；Initialize the BP neural network hidden layer weight coefficient, Output layer weight coefficient；
    Step (2)：The k moment will be denoted as current time；
    Step (3)：Actual value is exported with system based on system output desired value, calculating function using systematic error is calculated k The systematic error at moment, is denoted as e (k)；By the systematic error and its group of functions, system output desired value, system output reality One of the arbitrary or any number of combination of value, is put into set { systematic error collection }；
    Step (4)：The set { systematic error collection } that step (3) is obtained is as the input of BP neural network, the BP god Forward calculation is carried out through network, result of calculation is exported by the output layer of the BP neural network, obtains the inclined forms of the MISO Non-Model Controller treats the value of setting parameter；
    Step (5)：The inclined forms of the MISO that the systematic error e (k) that is obtained based on step (3), step (4) are obtained are without mould Type controller treats the value of setting parameter, and using the control algolithm of the inclined form Non-Model Controllers of MISO, the inclined lattice of MISO are calculated Formula Non-Model Controller is directed to control input vector u (k)=[u of the controlled device at the k moment1(k),…,um(k)]T；
    Step (6)：J-th of control input u in the control input vector u (k) obtained for step (5)j(k)(1≤j≤ M), j-th of the control input u is calculatedj(k) setting parameter is treated for each inclined form Non-Model Controllers of MISO respectively In the gradient information at k moment, specific formula for calculation is as follows：
    When the inclined form Non-Model Controllers of the MISO are when penalty factor λ is included in setting parameter, j-th of the control inputs uj(k) it is for gradient informations of the penalty factor λ at the k moment：
    When the inclined form Non-Model Controllers of the MISO are treated to include step factor ρ in setting parameter1When, j-th of the control input uj(k) for the step factor ρ1It is in the gradient information at k moment：
    When the inclined form Non-Model Controllers of the MISO are treated to include step factor ρ in setting parameteriAnd during 2≤i≤L, the jth A control input uj(k) for the step factor ρiIt is in the gradient information at k moment：
    Wherein, Δ uj(k)=uj(k)-uj(k-1),The row matrix of MISO system puppet piecemeal gradient estimated value for the k moment,For row matrixI-th piece of row matrix (i=1 ..., L),For row matrixJ-th of gradient component Estimated value,For row matrix2 norms；
    The set of the above-mentioned whole gradient information is denoted as { gradient information j }, is put into set { gradient information collection }；
    Other m-1 control input in the control input vector u (k) obtained for step (5), repeats this step Suddenly, until the set { gradient information collection } includes the set of all { { gradient information 1 } ..., { gradient information m } }, Ran Houjin Enter step (7)；
    Step (7)：Target is minimised as with the value of system error function, using gradient descent method, the institute obtained with reference to step (6) Set { gradient information collection } is stated, carries out systematic error backpropagation calculating, updates hidden layer weight coefficient, the output of BP neural network Layer weight coefficient, as later moment in time BP neural network carry out forward calculation when hidden layer weight coefficient, output layer weight coefficient；
    Step (8)：After the control input vector u (k) acts on controlled device, system of the controlled device in later moment in time is obtained Actual value is exported, back to step (2), repeats step (2) to step (8).
  2. 2. inclined methods of self-tuning of the form Non-Model Controller based on systematic error of MISO according to claim 1, It is characterized in that, the systematic error in the step (3), which calculates argument of function, includes system output desired value with being System output actual value.
  3. 3. inclined parameter self-tuning sides of the form Non-Model Controller based on systematic error of MISO according to claim 1 or 2 Method, which is characterized in that the systematic error in the step (3) calculates function and uses e (k)=y*(k)-y (k), wherein y* (k) system set for the k moment exports desired value, and y (k) is that the system that k instance samples obtain exports actual value；Or using e (k)=y*(k+1)-y (k), wherein y*(k+1) desired value is exported for the system at k+1 moment, y (k) is for what k instance samples obtained System output actual value.
  4. 4. inclined methods of self-tuning of the form Non-Model Controller based on systematic error of MISO according to claim 1, It is characterized in that, the systematic error and its group of functions in the step (3), when the systematic error e (k) comprising the k moment, k It carves and the accumulation of the systematic error at all moment before isThe single order backward difference e (k) of k moment systematic error e (k)- E (k-1), second order backward difference e (k) -2e (k-1)+e (k-2) of k moment systematic error e (k), k moment systematic error e (k) One of the arbitrary or any number of combination of high-order backward difference.
  5. 5. inclined methods of self-tuning of the form Non-Model Controller based on systematic error of MISO according to claim 1, It is characterized in that, the independent variable of the system error function in the step (7) includes systematic error, system output it is expected One of the arbitrary or any number of combination of value, system output actual value.
  6. 6. inclined parameter self-tuning sides of the form Non-Model Controller based on systematic error of MISO according to claim 1 or 5 Method, which is characterized in that the system error function in the step (7) isWherein, e (k) is Systematic error, Δ uj(k)=uj(k)-uj(k-1), bjTo be greater than or equal to 0 constant, 1≤j≤m.
CN201711323412.9A 2017-12-12 2017-12-12 System error-based parameter self-tuning method for MISO partial-format model-free controller Active CN108181808B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711323412.9A CN108181808B (en) 2017-12-12 2017-12-12 System error-based parameter self-tuning method for MISO partial-format model-free controller

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711323412.9A CN108181808B (en) 2017-12-12 2017-12-12 System error-based parameter self-tuning method for MISO partial-format model-free controller

Publications (2)

Publication Number Publication Date
CN108181808A true CN108181808A (en) 2018-06-19
CN108181808B CN108181808B (en) 2020-06-09

Family

ID=62546138

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711323412.9A Active CN108181808B (en) 2017-12-12 2017-12-12 System error-based parameter self-tuning method for MISO partial-format model-free controller

Country Status (1)

Country Link
CN (1) CN108181808B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111522229A (en) * 2019-02-01 2020-08-11 浙江大学 Parameter self-tuning MIMO different-factor offset format model-free control method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1274435A (en) * 1997-10-06 2000-11-22 美国通控集团公司 Model-free adaptive process control
CN101349893A (en) * 2007-07-18 2009-01-21 太极光控制软件（北京）有限公司 Forecast control device of adaptive model
CN101968629A (en) * 2010-10-19 2011-02-09 天津理工大学 PID (Proportional Integral Derivative) control method for elastic integral BP neural network based on RBF (Radial Basis Function) identification
CN107023825A (en) * 2016-08-31 2017-08-08 西安艾贝尔科技发展有限公司 Fluidized-bed combustion boiler is controlled and combustion optimizing system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1274435A (en) * 1997-10-06 2000-11-22 美国通控集团公司 Model-free adaptive process control
CN101349893A (en) * 2007-07-18 2009-01-21 太极光控制软件（北京）有限公司 Forecast control device of adaptive model
CN101968629A (en) * 2010-10-19 2011-02-09 天津理工大学 PID (Proportional Integral Derivative) control method for elastic integral BP neural network based on RBF (Radial Basis Function) identification
CN107023825A (en) * 2016-08-31 2017-08-08 西安艾贝尔科技发展有限公司 Fluidized-bed combustion boiler is controlled and combustion optimizing system

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
CHEN CHEN 等: "Design of Self-Tuning SISO Partial-Form Model-Free Adaptive Controller for Vapor-Compression Refrigeration System", 《IEEE ACCESS》 *
CHEN CHEN 等: "Parameter Self-Tuning of SISO Compact-Form Model-Free Adaptive Controller based on Neural Network with System Error Set as Input", 《2019 12TH ASIAN CONTROL CONFERENCE》 *
MINGWANG ZHAO: "Neural-net-based model-free self-tuning controller with on-line self-learning ability for industrial furnace", 《1994 PROCEEDINGS OF IEEE INTERNATIONAL CONFERENCE ON CONTROL AND APPLICATIONS》 *
李雪园: "无模型控制器参数在线自整定研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 *
郭代银: "无模型自适应控制参数整定方法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 *
马平 等: "无模型控制器参数学习步长和惩罚因子的整定研究", 《仪器仪表学报》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111522229A (en) * 2019-02-01 2020-08-11 浙江大学 Parameter self-tuning MIMO different-factor offset format model-free control method
CN111522229B (en) * 2019-02-01 2023-12-08 浙江大学 Parameter self-tuning MIMO different factor offset format model-free control method

Also Published As

Publication number Publication date
CN108181808B (en) 2020-06-09

Similar Documents

Publication Publication Date Title
CN108287471A (en) Inclined methods of self-tuning of the format Non-Model Controller based on systematic error of MIMO
CN108170029A (en) Methods of self-tuning of the MIMO full format Non-Model Controller based on local derviation information
CN108181809A (en) Tight methods of self-tuning of the form Non-Model Controller based on systematic error of MISO
CN110824923A (en) Sewage treatment control method and system based on deep learning and cloud computing
CN108153151A (en) Methods of self-tuning of the MIMO full format Non-Model Controller based on systematic error
CN109581864A (en) The inclined format non-model control method of the different factor of the MIMO of parameter self-tuning
CN108345213A (en) Tight methods of self-tuning of the format Non-Model Controller based on systematic error of MIMO
CN109254530A (en) MFA control method based on grinding process basis circuit
CN105469142A (en) Neural network increment-type feedforward algorithm based on sample increment driving
CN107942655A (en) Tight methods of self-tuning of the form Non-Model Controller based on systematic error of SISO
CN108154231A (en) Methods of self-tuning of the MISO full format Non-Model Controller based on systematic error
CN108062021A (en) Methods of self-tuning of the SISO full format Non-Model Controller based on local derviation information
CN108181808A (en) Inclined methods of self-tuning of the form Non-Model Controller based on systematic error of MISO
CN107942654A (en) Inclined methods of self-tuning of the form Non-Model Controller based on local derviation information of SISO
CN108132600A (en) Tight methods of self-tuning of the form Non-Model Controller based on local derviation information of MIMO
CN108073072A (en) Tight methods of self-tuning of the form Non-Model Controller based on local derviation information of SISO
CN108008634A (en) Inclined methods of self-tuning of the form Non-Model Controller based on local derviation information of MISO
CN108107715A (en) Methods of self-tuning of the MISO full format Non-Model Controller based on local derviation information
CN109782586A (en) The tight format non-model control method of the different factor of the MISO of parameter self-tuning
CN108287470A (en) Inclined methods of self-tuning of the format Non-Model Controller based on local derviation information of MIMO
CN108052006A (en) Decoupling control methods of the MIMO based on SISO full format Non-Model Controller Yu local derviation information
CN108388115A (en) NCS method for compensating network delay based on generalized predictive control
CN111539508A (en) Generator excitation system parameter identification algorithm based on improved wolf algorithm
CN108107727A (en) Tight methods of self-tuning of the form Non-Model Controller based on local derviation information of MISO
CN107942656A (en) Inclined methods of self-tuning of the form Non-Model Controller based on systematic error of SISO

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant